Beginning the Search for the Inverse Fission of Uranium

Transcription

1 Beginning the Search for the Inverse Fission of Uranium

2 AN ABSTRACT OF THE THESIS OF John R. Beckerman for the degree of Master of Science in Chemistry presented on June 9, Title: Beginning the Search for the Inverse Fission of Uranium Abstract approved: Walter D. Loveland The nuclear reaction 100 Mo( 124 Sn,xn) 224-x U has been carried out at E cot of 578 MeV and 634 MeV using the facilities and equipment at the Holifield Radioactive Ion Beam Facility (HRIBF) at Oak Ridge National Laboratory (ORNL) in order to initiate the search for the inverse fission of uranium. Reference beams of 238 U at 331 MeV and 365 MeV were also used to aid in the search for possible evaporation residues. A bismuth target was employed to attempt to obtain a background spectrum to more accurately determine the overall cross section, which was deterimined to be 0.2 mb at E cot of 578MeV and 0.5 mb at E cot of 634 MeV. A discussion of the experimental apparatus, techniques used to analyze the data, as well as the use of the bismuth target as a background is included.

3 Copyright by John R. Beckerman June 9, 2010 All Rights Reserved

4 Beginning the Search for the Inverse Fission of Uranium by John R. Beckerman A THESIS submitted to Oregon State University In partial fulfillment of the requirements for the degree of Master of Science Presented on June 9, 2010 Commencement June 2011

5 Master of Science thesis of John R. Beckerman presented on June 9, APPROVED: Major Professor, representing Chemistry Chair of the Department of Chemistry Dean of the Graduate School I understand that my thesis will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my thesis to any reader upon request. John R. Beckerman, Author

6 ACKNOWLEDGEMENTS Many thanks are given to everybody involved in this study; Walter Loveland, Ricardo Yanez, Dan Shapira, Felix Liang, Carl Gross, and Robert Varner.

7 TABLE OF CONTENTS Introduction... 2 Page Literature Review Materials and Methods Data Analysis Results and Discussion Future Work Bibliography Appendices Appendix A: Isolation and Molecular Plating of Uranium... 56

8 LIST OF FIGURES Figure Page 1. Nuclear binding energy per nucleon as a function of mass number Calculated one-dimensional potential between two 90 Zr atoms as a function of distance Measured evaporation residue cross sections for the 100 Mo+ 100 Mo, 100 Mo+ 110 Pd, and 110 Pd+ 110 Pd reactions This is the detection system used Schematic of the MCPs used in the experiment An outline of the beam production facilities at ORNL Signal processing diagram of the data acquisition system Raw two dimensional spectra of the 595 MeV run in the reaction of 124 Sn+ 100 Mo The energy loss spectrum of 331 MeV uranium with the gates shown as the black lines The time of flight spectrum, labeled tof1 in the analysis code, from MCP1 to MCP2 using 650 MeV 124 Sn Results from the 595 MeV run Results from the 650 MeV run Decay Scheme for 230 U Basic electronics diagram for the alpha spectrometer

9 LIST OF TABLES Table Page 1. A summary of some possible candidate reactions to form 230 U.. 46

10

11 Beginning the Search for the Inverse Fission of Uranium

12 2 Introduction In the field of nuclear science, the study of heavy elements constitutes a very large body of research. The heaviest elements are proving grounds for testing new theories of nuclear structure and interactions. One longstanding phenomenon characterizing heavy elements is of nuclear fission and fusion. Fission was first observed in ,2 while investigating the possibilities of forming a new and heavier element. Instead of seeing something heavier, fragments that were much smaller were found, and the idea of fission was born. Ever since, the complete understanding and modeling of the fission process has been sought after, and still remains elusive today. Undoubtedly, knowing more about the process of fission is of great interest to the nuclear power industry, where fission and its byproducts are ubiquitous. Fusion using hydrogen may even have a place in nuclear energy sources in the future, since this process also releases large amounts of energy. By being able to better understand both fusion and fission, the process of both can be modeled more effectively. Better models and systematics mean better representations of nuclear potentials and barriers, which can give rise to more efficient energy systems. In addition to benefitting industry, clearly understanding fission can provide new channels of research. Many cutting edge experiments require the use of exotic beams to carry out the desired reaction. Hence, if it can be accurately predicted what fragments can be expected from the fission of a given isotope, new exotic beams can be produced to meet the needs of novel research. However, for any of this to be plausible, one must back up to the first step of clearly understanding the process of fission. Part of the difficulty of understanding fission lies in how

13 3 we observe it experimentally. A typical experiment involves irradiating a target with a projectile that is expected to induce fission, and then recording the results. In this way, one can know the start and the end, but have to rely on theories to fill in the middle, which is where the actual process of fission occurs. As a result, it has been suggested that it could be beneficial to attempt to run the process backwards. 3,4,5 This is what is meant by the term Inverse Fission. Rigorously defined, however, inverse fission is not exactly the process of fission in reverse. In fission, the fragments created are in highly excited states, which are too difficult to produce in a laboratory environment. Instead, inverse fission utilizes nuclei with proton numbers similar to the fragments produced in fission, but which aren t in highly excited states. Although inverse fission can t mimic the excited states of the fission fragments, it carries with it the advantage of being able to select a wide range of combinations of possible projectiles and targets. As a result, the process of inverse fission is relatively symmetric compared to most other reactions that produce heavy nuclei. If it can be observed how the process of fission behaves in the reverse direction, it may be possible to have a glimpse into the large scale collective motion present during the splitting of a nucleus. Most importantly, inverse fission is, in essence, a fusion reaction testing the limits of heavy ion fusion that can possibly cast light on the feasibility of the fusion of symmetric systems. As with fission, fusion of heavy systems is, and has always been, an intellectual challenge as well. When a consistent description of various reaction processes is used, deviations of the experimental from the predicted measurements are observed. Such is found when implementing the same framework for the fusion of systems ranging from far below the fusion barrier to far above the fusion barrier. 6 As a result, a single, reliable, framework would be highly advantageous, and inverse fission reactions have the potential to test these new models.

14 4 The first fusion reactions leading to new heavy elements were carried out by bombarding a target, such as uranium, with neutrons and observing the new nuclides present after ensuing β - decays. One must keep in mind when selecting reactions, the balance between the attractive nuclear potential and the opposing coulomb repulsion. The coulomb interaction between two nuclei can be described by the following equation: EE CC = ZZ 1 ZZ 2 ee 2 RR where ZZ 1 and ZZ 2 are the number of protons, ee 2 is the elementary charge, and RR is the distance separating the two nuclei. Two things should be noted from the equation; first is that as two nuclei come closer together, there will be more repulsion, and second, the larger the nuclei, the larger the repulsion. This logic is why neutrons were the first candidates to be used in reactions involving heavy nuclides; the neutrons experience no coulomb repulsion since they do not carry a charge. They merely must be captured by the nuclear potential. This was behind the synthesis of the first element beyond uranium: 238 UU nn,γγ 239 UU ββ 239 NNNN. This manner of bombardment with neutrons is successful to produce nuclides up to Z=100. At this point, neutron capture leading to higher Z nuclei stops due to the extremely short half-life of 258 Fm. Its half-life of 0.4 milliseconds effectively prevents further buildup of any possibly useful new nuclei. However, as technology advanced, so did the periodic table of the elements. When cyclotrons became available, it suddenly was possible to overcome the coulomb barrier and collide large projectiles, such as protons and alpha particles, with heavy elements. When analyzing these nuclear reactions, it is useful to look at fusion as a two-step process. The first step is the capture of projectile and target due to their nuclear attraction, but only if there is sufficient energy to overcome their mutual coulomb repulsion. At this point, a compound

15 5 nucleus consisting of the target and projectile situated as two spheres just touching is formed. If this compound nucleus goes on to form a compact compound nucleus, where target and projectile are no longer distinguishable, then fusion can be said to have taken place. The following equation represents these processes involved in fusion: σσ ffffff = σσ cccccc PP CCCC where σσ ffffff represents the fusion cross section, σσ cccccc is the capture cross section, and PP CCCC is the probability that the compound nucleus will evolve from the contact configuration to form a compact compound nucleus. However, there is one more step that should be added. Presuming that fusion has taken place, the resulting compound nucleus will likely be in an excited state. If this excited state is above the fission barrier, then there are competing pathways to the dissipation of this excitation energy. Of course in order for this new nucleus to be studied, it must dissipate its energy in ways other than fission. This can be reflected in the above equation by amending an additional term, WW ssssss, which is the likelihood of the compound nucleus to survive the subsequent statistical fission decay. 7 If all of this happens, an evaporation residue is formed according to this equation, with σσ EEEE denoting the evaporation residue cross section: σσ EEEE = σσ cccccc PP CCCC WW ssssss A few basic things can be learned from the above equation. Since σσ cccccc is dependent on the coulombic repulsion of the reacting ions, we can readily see that target and projectile combinations with large values of ZZ 1 *ZZ 2 will be much harder to utilize. If we force two heavier ions to react by using more powerful accelerators and cyclotrons, then we also start killing the yield through the WW ssssss term. Once in the realm of forming the very heaviest nuclei, close

16 6 attention must be paid to the properties of the reacting nuclei. This is where the differences between the two synthetic procedures of hot and cold fusion can be seen. When attempting to make the heaviest elements, larger nuclei must not only be used as targets, but the projectiles will also be necessarily large. In hot fusion, large target nuclei, like the actinides, are collided with the corresponding projectile to make the desired new nuclide. These projectiles can range from 11 B through 48 Ca. A good example of hot fusion is the one used to make Seaborgium (Sg): 249 CCCC 18 + OO 263 SSSS + 4nn There are, however consequences to the use of heavier target and projectile combinations. More energy is required to initiate the reaction due to the increasing coulomb repulsion. The resulting compound nucleus is left in a highly excited state. A trend among the limits of the periodic table is the increasing fissility of the heaviest nuclei as the proton number increases. This means that the fission barrier can be more and more easily overcome, and it follows that any nuclei with an appreciable amount of excitation energy will be likely to undergo fission. Typical excitation energies in hot fusion reactions are around 30MeV to 50MeV. This is reflected in the half lives of the heaviest nuclei: fermium is 20 hours, lawrencium is 4.3 seconds, rutherfordium is 3.8 seconds, and in the reaction above, seaborgium is only 0.9 seconds. Beyond seaborgium, the survival probabilities of new nuclei from hot fusion become very poor. The nuclear physicist Yuri Oganessian suggested an alternative pathway to making heavy nuclei by using targets in the lead-bismuth region. Lead and bismuth have characteristics that make them quite stable, which can give rise to compound nuclei with lower excitation

17 energies, hence the name cold fusion. The following reaction producing roentgenium (Rg) is a good example: BBBB 64 + NNNN 272 RRRR + nn In cold fusion reactions like the one above, excitation energies are around 13MeV. This means a lower chance of fission, and is quite an improvement from the excitation energies involved in hot fusion. As with hot fusion, however, there are also drawbacks of cold fusion. Most importantly is the increased repulsion between target and projectile because of the combined larger proton numbers in each. This translates to a lower fusion probability. And again the trend of increasing instability towards the end of the periodic table is evident among cold fusion as well, with half lives in the range of milliseconds, and decreasing into the microsecond range. For reference, in the above reaction giving rise to an atom of Rg, it only has a half life of 1.5 ms. It must always be kept in mind that fusion is only one of the outcomes of these reactions. We have already seen the damaging effects that fission has on the evaporation residue cross section, as well as how it has been attempted to work around. There are additional outcomes other than fusion and fission that are worth noting. By far the most common outcome is that of elastic scattering reactions. These reactions include Coulomb scattering, which is commonly referred to as Rutherford scattering because of his famous experiment with alpha particles and a gold foil, and compound elastic scattering. Coulomb scattering is a long range deflection process due to the electrostatic repulsion of the collision partners, and it is important to note this does not actually involve the physical interaction of two nuclei. Compound elastic scattering is an interaction where the collision partners come into contact, but the projectile is

18 8 emitted with the same energy, resulting in a process that is often compared to two pool balls hitting each other. However, the other types of reactions are a result from varying amounts of interactions when the projectile hits the target. When the projectile comes into contact with the target, it forms the aforementioned contact configuration. From here it can undergo quasi-fission, deep inelastic scattering, fusionfission, or form a new compound nucleus. In quasi-fission, the target and projectile form an elongated dinucleus and then proceed to fission from this deformed shape. The fragments produced here can have very different angular and mass distributions than fragments from the fission of a compact compound nucleus. 8,9 In deep inelastic scattering the target and projectile partially fuse and then separate. During the time of contact, however, there is an exchange of mass and charge as well as the dissipation of a large amount of kinetic energy, which can also give rise to unique residual fragments. Fusion-fission is just the process of fission from a compact compound nucleus. In the end, a new compound nucleus is what is sought, and when choosing a reaction, it is desirable to minimize all of these other possible outcomes as much as possible. Unfortunately, this is not an easy task. When examining the kinetics of a given reaction, it can be calculated what the coulomb barrier should be, and hence find the amount of energy required for the nuclei to come into contact. Merely getting to this point is often not enough. It may be more preferable to provide enough energy such that upon collision, the conditional saddle point can be overcome. This conditional saddle point is when the elongated dinucleus is formed. The amount of energy required to achieve this configuration is referred to as the extra push energy. This can be taken one step farther as well. In some cases, experimenters may decide to

19 9 provide enough kinetic energy that the unconditional saddle point will be overcome. Once this barrier is passed, a compound nucleus is formed. This amount of energy to overcome the unconditional saddle point, as may be guessed, is called the extra-extra push energy, although popular nomenclature has replaced this term with the more convenient one mentioned above, the extra push energy. The extra push energy can be problematic, however. Often times a disagreement arises between actual results and those calculated using the extra push energy, because, after all, the calculation of the extra push energy has its roots in theoretical and semiempirical equations. This leads to studies of where these discrepancies come from, and a large contributor is from entrance effects such as the internal nuclear structure of the target and projectile themselves. Much research has been done on the shell effects of nuclei. The result is that there are certain numbers of nucleons that are more stable than others, very much like the shell closures of orbitals, s and p for example, in molecules that depend on the correct number of electrons. These stable configurations consist of the number of nucleons, either protons or neutrons, which add up to 2, 8, 20, 28, 50, 82, or 126. Nuclei that exhibit one of these magic numbers display more stability than others. This can be seen in figure 1, which is the nuclear binding energy per nucleon. The small peaks, as well as the overall peak which lies at Fe, are due to extra stability from these magic numbers. These are the characteristics talked about when using lead or bismuth in cold fusion; 208 Pb has 82 protons and both 208 Pb and 209 Bi has 126 neutrons. Another characteristic that this property of magic numbers leads to is the shape of nuclei. If it has closed shells it will be much

20 10 Figure 1 Nuclear binding energy per nucleon as a function of mass number. The small peaks are attributed to shell closures. The black line is experimental result, while the red line is from the semiempirical mass equation. 10 more spherical. Generally speaking, spherical nuclei with closed shells will have lower excitation energies, which give rise to larger values of WW ssssss. But, it may not always be easier to bring these nuclei into contact. More energy is needed to bring two positively charged spheres together relative to distorted spheres which can touch at a greater distance where the repulsion is less. This means that the overall nuclear structure of the projectile and target combination is related to the touching probability, but it also influences the dynamics from the touching point to the compound nucleus. 11,12 When nuclear structure is taken into account and shell effects are considered, cold fusion valleys are observed in the potential energy surface, and these valleys are what give rise to the probability of a compound nucleus. 13

21 11 The area of producing the heaviest nuclei is where inverse fission could have its place in the distant future. The elements from nobelium to seaborgium were first made using hot fusion. Past these elements, from bohrium to element 113, cold fusion was utilized. As one goes up on the periodic table, it becomes increasingly difficult to produce nuclei using either approach, but it has been predicted that hot fusion will give higher cross sections than cold fusion. 14 Inverse fission provides a possible third pathway to making heavy nuclei, although much research would have to be done. It would be ideal to start by trying to produce the lighter nuclei in the actinide series, and from there investigate the feasibility to apply the process of inverse fission to heavier and heavier systems. The aim of this work is to begin the search for inverse fission. One of the most common elements used today that undergoes fission is uranium, and it is also the heaviest naturally occurring element on the periodic table. This would make uranium an, albeit arbitrary, but also convenient, place to initiate this search. Perhaps a more determining feature for beginning with uranium than its place on the periodic table is the availability of components that could be used to make uranium. The primary source of fission fragments from uranium is from 235 U, although 238 U can be made to fission as well, so when it fissions two fragments are produced that have mass numbers around 90 and 140. When looking for possible combinations of targets and projectiles in this region, the use of tin is appealing. Tin (Sn) has 50 protons, which is a magic number, and has a stable isotope with a mass number of 124, which is similar to a fragment produced in fission. Because of the research already completed on fission, there exists a number of exotic beams that are available to experimenters. One of these beams is of great interest to the

22 12 inverse fission of uranium: 132 Sn. This isotope is doubly magic, meaning it has 50 protons and 82 neutrons, so all of its nucleons lie in closed shells. This is a neutron/proton ratio of 1.64, which is very neutron rich compared to the rest of the periodic table, which has N/Z ratios starting at 1 and going to 1.5 among the heavy elements. Like any nucleus that is neutron rich, however, it will undergo β - decay. In this case, 132 Sn has a half life of 39.7 seconds. Producing this beam can be problematic, however, so as a first step the stable isotope 124 Sn will be used. The proposed reaction is that of 100 Mo ( 124 Sn,xn) 224-x U. A number of characteristics can be described about this reaction that relate to what was discussed above. Of course the most notable trait is the magic nature of the 124 Sn nucleus, but it is also worth examining the structure of 100 Mo as well. Being used as the target, 100 Mo has 42 protons and 58 neutrons which make it sit in-between major shell closures. As a result the nucleus will be slightly distorted, but this is not necessarily a bad thing. Because of its deformation, it could allow interaction with 124 Sn from a farther distance, which can enhance the contact cross section. If all goes well, the compound nucleus will form with a lower excitation energy, and an evaporation residue will result. There is still a problem with supplying the correct amount of energy to ensure interaction between 124 Sn and 100 Mo. Since ZZ SSSS *ZZ MMMM is equal to 2100, there is considerable repulsion, and more repulsion usually translates to lower cross sections. This can be compared to ZZ BBBB *ZZ CCCC for the reaction 209 Bi( 54 Cr,n) 262 Bh which has a value of Bohrium is one of the heaviest atoms having 107 protons and a lifetime of 102 ms, so it can be seen what difficulties must be surmounted for a successful reaction of the inverse fission of uranium. Experimenters can become limited by the facilities which can provide the requisite beams for particular reactions. In this case, Oak Ridge National Laboratory has the means to

23 13 provide Sn beams, including 132 Sn. Predictions, however, can often be optimistic when dealing with such specialized beam requirements. That is why a 124 Sn beam is being used as a first step; it is far easier to produce, much more reliable, and also available. This reaction will also give a general idea of the feasibility of inverse fission as well as insight into the fine balance of all the forces at play during a relatively symmetrical nuclear reaction.

24 14 Literature review Inverse fission of uranium has not had much attention in the research community. There is a wealth of research concerning symmetric fusion partners, but very few are related to uranium. As an example of the barriers involved in symmetric fusion, consider figure 2. This represents the electrostatic and proximity potential, which is a description of the nuclear potential according to Blocki et al 15, for two 90 Zr nuclei. For comparative purposes, something similar could be expected for the reaction presented in this paper. The 90 Zr( 90 Zr,γ) 180 Hg reaction is in some ways similar to the 124 Sn reaction. The reaction partners are stabilized by a closed shell of neutrons, and there is also a large, albeit somewhat less than the 124 Sn reaction, coulomb repulsion interaction. The closed shell favors a lower excitation energy, while the fission barrier of 180 Hg is high enough to limit the competition of fission in de-excitation. The Figure 2 Calculated one-dimensional potential between two 90 Zr atoms as a function of distance. 15

25 15 fusion of two 90 Zr nuclei lead to the discovery of a new reaction mechanism between heavy nuclei, that of radiative fusion. This means that the dissipation of the excitation energy was only by releasing γ-rays, and not evaporating any nucleons. This was confirmed by the α-decay of the ground state evaporation residues. 16 As the reaction system becomes heavier and heavier, fusion barriers start to change. Calculations of fusion cross sections, as calculated from the penetration of a one-dimensional potential, as the one above, start to stray from experimental values. This is an indicator that one is at the threshold of fusion hindrance, and it also suggests that there are other forces at play besides the one-dimensional potential. As mentioned earlier, this is where the extra push energy comes into play. This can be seen as a decrease in evaporation residue cross section as one goes to heavier symmetric fusion reaction, such as fusing two 110 Pd nuclei. The reaction of fusing two 110 Pd is of interest here because it has the possibility of forming an atom of 220 U. 17 It represents one of the only real attempts at the inverse fission of uranium. Figure 3 contains the measured evaporation residue cross sections for this experiment, as well as that of some other related and symmetric reactions. Although it can be seen that the two 110 Pd nuclei fused, the cross section is on the order of nanobarns. This is a result from the combination of fusion hindrance and a strong competition from the fission de-excitation channel. Also, the gradual increase in the evaporation residue cross-section is a telltale signature of a very broad effective fusion-barrier distribution. 18 Much of the fission competition could be attributed to the compound nucleus formed. Since 220 U is so neutron deficient, it is very fissionable. After all, the chart of the nuclides lists 220 U as having a half-life of roughly 60 nanoseconds. Altogether, this doesn t lead to any convincing evidence for inverse fission.

26 16 Figure 3 Measured evaporation residue cross sections for the 100 Mo+ 100 Mo, 100 Mo+ 110 Pd, and 110 Pd+ 110 Pd reactions. The arrows point to the one-dimensional potential barrier in each reaction. 19 The inadequacy of the one-dimensional barrier has led to using new models, such as the coupled channels approach and the dinuclear system model (DNS), to analyze heavy ion reactions. Consider the reactions 124 Sn + 96 Zr, and 132 Sn + 96 Zr, where the capture-fission excitation functions were used to further understand the entrance channel dynamics of using neutron rich projectiles. 20,21 In these reactions, the coupled channels approach is more cumbersome to use; it, in fact, did not even work, so the DNS system is preferred. In both cases, however, the one-dimensional model falls quite short of accurately reproducing the data. Another interesting note to make about this reaction is the effect neutron richness has on the capture fission cross section. Normally it is expected that using more neutron-rich collision

27 17 partners will give rise to higher cross sections, but the results obtained were somewhat contradictory. In the 124 Sn Mo reaction, it will also be of interest what the effect neutron rich projectiles will have, such as 132 Sn. The experimental analysis will be slightly different than that of the 124,132 Sn + 96 Zr reactions. Here, one is looking at the evaporation residue cross sections instead of extrapolating a fusion cross section from a capture-fission excitation function, as was done in references [20],[21]. Recall that the evaporation residue cross section includes the decay of the compound nucleus, while the fusion cross section takes into account all steps up to compound nucleus formation but does not include its decay. By considering the ER cross section, effects of neutron richness can be observed in the formation of heavy residues. There is another reaction worth noting that has led to the formation of uranium atoms, that of 180 Hf ( 48 Ca,3n) 225 U. 22 This reaction was carried out at a single energy, and a cross section of 130 nb was obtained. Such a low cross section suggests that perhaps the energy chosen was not optimal since other similar reactions, such as 48 Ca Pb, show much higher yields. This reaction is also rather asymmetric, which is not the aim of inverse fission. Nonetheless, it shows that there have been successful attempts to making uranium. In conclusion, if the ideas obtained from these previous reactions are extrapolated to this work, many difficulties can be seen. Foremost is that of coulomb repulsion, and the tradeoff of extra push energy and excitation energies. Current theories and models have been relied upon to give the best estimate of optimal bombarding energies and anticipated kinematics. It follows that the results obtained here will give some insight into the functionality of these models.

28 18 Materials and Methods The reaction of 100 Mo( 124 Sn,xn) 224-x U was carried out using the Holifield Radioactive Ion Beam Facility (HRIBF) at Oak Ridge National Laboratory (ORNL), located in Oak Ridge Tennessee. A high-efficiency system for detecting heavy ion residues has been developed at ORNL, and this system is what was used to investigate this reaction. 23 A schematic of the apparatus can be seen in figure 4. Three timing detectors are used in all; two are preceding the target and one comes after the target. After the third timing detector is the ionization chamber, which is used to measure the energy loss of all incoming particles. Figure 4 This is the detection system used. The blue boxes represent parameters that are recorded every time a valid event is registered. A value of 1000 is used for n. The distances are as follows: T0-T1=813mm; T1-Taget=1130mm; Target-T2=169mm; T2-IC=13mm. 23

29 19 To obtain satisfactory timing signals, microchannel plate detectors (MCPs) are used, which represent T0, T1, and T2 in figure 4. These detectors consist of a thin carbon foil, a reflecting grid, an MCP assembly, and a resistive layer or anode plane. When a beam particle or residue strikes the carbon foil, electrons are ejected. The carbon foil is held at roughly volts, so any electrons produced are rapidly repelled to the reflecting grid. The reflecting grid is insulated by a grounded grid plane, and is held at a negative potential of about 1500 V on the first two MCPs and about 1700 V on the second. The electrons approach the grounded grid plane easily, but after passing through, immediately experience the negative potential of the reflecting grid and are bent 90 due to the angle of the reflecting grid. This assembly can be thought of as an electrostatic mirror that redirects the ejected electrons. After being reflected, the electrons then enter the MCP assembly. This assembly is comprised of two microchannel plates which greatly amplify the number of electrons. These plates work in similar ways to photomultiplier tubes, but instead of having several dynodes, more electrons are produced by striking the inside walls of tiny channels in the plate. These tiny channels are slanted, such that any incoming electrons will come into contact with the interior of the channel. On a larger scale these microchannel plates could be simplified by a comparison to a bundle of slanted straws, with the interior surfaces of these straws acting as the source of electron multiplication. After passing through the MCP assembly, the electrons are gathered and a signal is produced by either a resistive layer or a metal anode. The anodes are held at a positive voltage of around 2100 V, and if a resistive layer is present, it is at around 2250 V. A more in depth discussion of the MCPs can be found elsewhere 24, but for a graphical representation, see figure 5. Using these MCPs is critical to obtaining good measurements. During the experiment, the intensity of the beam is held anywhere from 50,000 to 100,000 particles per second. For this

30 20 Figure 5 Schematic of the MCPs used in the experiment. The first two MCPs are only timing detectors, while the third can also provide information on the position of the beam. The voltages indicated are generalized; see text for the specific voltages used. 23 range of intensities, very good timing resolution is required. Microchannel plate detectors are ideal for this, since 100 ps timing resolution can be achieved at count rates in excess of 1MHz. The third MCP also has capabilities of acting as a position detector as well. Instead of a metal anode like the other MCPs, this one has a resistive layer. The charges collected at the four corners of this resistive layer are then used to determine the position of the electron. This is because the position of a signal at the exit of the MCP assembly will reflect the original position of where the beam or residue struck the foil. This will, of course, be accompanied by some uncertainty due to spreading of the electrons. The spatial resolution is expected to be around 2 mm, which is good enough to ensure the beam is being delivered to the center of the target. 23 As indicated in figure 4, the first two MCPs are used to monitor properties of the beam, as well as to generate an event trigger on the beam. This trigger is produced by adjusting the delay such that it will fire on particles having the correct velocity. This trigger is scaled down by a

31 21 factor of 1000 and can be combined with other parameters during the data analysis stages. Having this trigger allows calculation of the number of beam particles delivered, and therefore the cross section of the reaction. The second and third MCPs are also used in a similar fashion to generate another trigger on particles that move slower than the beam, which will include any evaporation residues. When either of these instances occurs, an event is said to have taken place. The overall result is a trigger rate of less than 1 khz, and no loss of any events of interest. After the series of timing detectors lies the ionization chamber. Here is where the energy loss of all incoming particles is measured. It lies only 13 mm from the last timing detector and has an entrance window which is 2.5 cm in diameter. This translates to being able to accept particles emanating in a 3.5 cone from the center of the target. This should be plenty of leeway since it is predicted that any residues produced in the target should be emitted in a 0-2 cone. This is because any particles that are expected to be emitted, i.e. one or more neutrons, will have a very small mass compared to the residue. When a particle is emitted, the residue will experience very little recoil and retain a somewhat straight trajectory, and in this case is within the spatial capabilities of the ionization chamber. The entrance window is constructed of a thin piece of Mylar, which is a strong polymer with excellent tensile strength and gas barrier properties to allow low pressures on one side, and a higher pressure inside the ionization chamber. The chamber is in total 30 cm long, but this distance is divided into three regions; the first two being 7.5 cm long and the last being 15 cm to measure the residual energy of stopped particles. The gas used in the chamber is carbon tetrafluoride, CF 4, which has good enough stopping capabilities and electron mobility to be able to use this chamber at beam intensities of greater than 50 khz. The pressure of the chamber is something that can change from experiment to experiment, depending on the incoming particles. The pressure should be low

32 22 enough to allow any evaporation residues to penetrate to the area of the third anode, but be high enough to prevent the residues from striking the end of the detector with excess energy. Since the pressure is directly related to the stopping power, the values of energy loss recorded at each anode will vary as well. This means that the pressure influences how well the energy loss of evaporation residues can be isolated and recognized from the rest of the beam particles. For this experiment, 35 torr provided optimal conditions. Other pressures of 20, 25, and 30 torr were tried, but did not give good separation between the area expected for uranium residues and from that of the beam. It must be pointed out that this detection system of using an ionization chamber which collects information on all events, which is overwhelmingly all beam particles, is only useful at intensities under 100 khz, which is generally thought of as low-intensity. If higher intensities are used, then problems arise because of dead times in the computer system and other areas of the electronics used for data acquisition. If higher intensities are sought, a different system must be used in which the beam is separated from the residues of interest. There are, however, advantages to both systems. By not separating the beam and measuring everything that comes into the chamber, as is done with this experiment, no events are lost and the efficiency is very high, in the range of 95% to 100%. However, this comes at the cost of having to wade through much more data to find the events of interest, which consequently means a higher random background rate. If an online-separation method is used, higher intensities can be used which results in being able to achieve lower cross sections in shorter timeframes, but can also suffer from lower efficiencies.

33 23 All particles that enter the ionization chamber will have a characteristic energy loss spectrum. If these particles come in under the same conditions, then peaks will readily start forming. If one then plots the energy loss measured at the first anode against the energy loss of the second or third anode, a two dimensional histogram is created that can be used to visually separate events of interest. Unfortunately, because of the drawbacks discussed above of this system, a raw two dimensional spectrum doesn t readily lend itself to any conclusions. Limitations on certain parameters must be imposed in order to successfully separate events of interest from any beam particles. This topic will be dealt with in detail in Chapter 4. When an event coincidence is satisfied, numerous parameters are recorded. These include the charges accumulated at the four corners of the resistive layer of the third MCP, the signals of all three timing detectors, the time of flight between the different timing detector pairs, the energy loss from each anode in the ionization chamber, the total energy loss, and a bit register which records which type of event it is; either a scaled down beam event or an evaporation residue event. The charges recorded on each corner of the resistive layer are used to determine the beam position to ensure that the target is being irradiated at the proper position, which is very useful to know so the beam can be retuned as needed. The other parameters are used to provide extra arguments for finding events of interest during data analysis. As can be guessed from the number of parameters recorded every time an event occurs, very large file sizes are common. A typical two-day run can easily produce 18 GB of data. Although not a primary topic, it is still worth while to explain the procedures and equipment used at ORNL to produce the 124 Sn beam that was used in this experiment. The use of the isotope separator on-line technique is employed at ORNL to provide radioactive beams; it

34 24 is summarized in figure 6. This method of isotope separation is where products from the ion source are ionized, formed into a beam, and then mass selected for input into a 25 MeV electrostatic tandem accelerator. This process is very similar when producing stable beams, but the difference lies in the ion source. The production of unstable beams relies on the operation of the Oak Ridge Isochronous Cyclotron (ORIC), which has the ability to accelerate light ions, such as protons, to high energies. These high energy protons are then focused on a uranium carbide target, which subsequently fissions and emits numerous fission fragments. From there the ISOL process takes place to choose the fission fragments of interest to produce the desired beam. Having the ability to select specific isotopes from the huge array of fission fragments is Figure 6 An outline of the beam production facilities at ORNL. In this experiment, ORIC was not needed. Instead of a uranium carbide target, a sputter source is used. 25

35 25 the reason why the HRIBF is so unique and can provide such exotic beams. For stable beams, it is slightly simplified. There is no longer a requirement for fission fragments, so ORIC is not needed. Instead, the beam is produced from a sputter source composed of the isotope to be used for the stable beam of interest. Referring to this source as a sputter source means that the beam is procured via a sputtering mechanism, which can be likened to ablation. Because the isotope is stable, very pure sources can be used, usually approaching 100%. Initially, when the beam is formed, all of the ions are positively charged, but the tandem accelerator requires the beam to be negatively charged; hence the use of a charge exchange cell. This cell provides a collision vapor thickness of about atoms/cm 2 to allow conversion of a positive beam to that of a negatively charged beam. Typical efficiencies here, for unstable beams, can be around 10%-50%. In addition, the beam is further purified by the isobar separator after passing through the charge exchange cell. This is important when delivering unstable beams, since the beam of interest is not always the major byproduct of fission. As an example, tellurium is initially the primary component of a mass 132 beam when producing 132 Sn, but must be removed to provide a high purity beam. Most of the same equipment is used when producing stable beams, although they do not require excessive purification techniques. After passing through the mass analyzer and charge exchange cell and isobar separator, the beam is negatively charged and around 200 kev in energy, and is now suitable for injection into the tandem accelerator. The 25 MeV tandem accelerator is, as of 2005, the world s highest voltage electrostatic accelerator. This accelerator is housed in a 100 foot tall, 33 foot diameter pressure vessel insulated with pure SF 6 gas. After passing through the accelerator, the beam is then delivered to the target at the desired beam energy. In this experiment, energies up to 650

36 26 MeV were obtained without much difficulty. As a whole, HRIBF can provide both neutron and proton rich beams found nowhere else in the world. In order to deliver these beams and have ideal conditions in the beam lines from the ion source to the target, however, pressures must be kept low; in the range of 10-7 torr. If one is using MCPs, the pressure must be on this scale, otherwise damage can be inflicted to the components of the MCP. Recall that the components of the MCP are held at high potentials, so if the pressure is high enough that it provides a channel for arcing, it can easily destroy an MCP. There are safety precautions, however. If something does arc or spark, the voltage is immediately terminated to that component, and the potential must be manually returned to the desired level. In addition, the pressure must be this low to ensure the passage of electrons through the assembly. This is of greater importance in the two-layer MCP assembly where particles can be easily trapped in the tiny channels. While this is not difficult to obtain a pressure on the scale of 10-7 torr in the beam lines leading up to the experimental area, it can provoke some problems in the area of the beam lines where the MCPs and target are located. The thin carbon foils of the MCPs are very thin, only 10 µg/cm 2. Being so fragile, clever procedures must be done so that the foils don t break from excess drag produced from pumping too quickly. Small diameter bypass pipes are used to initially pump through until the pressure is low enough to allow normal pumping and the use of high vacuum pumps. In this setup, there are two high-vacuum pumps. One is pumping on an area close to the first timing detector, and the second is a cryogenic pump that is pumping where the target, third timing detector, and ionization chamber is located. Once a safe procedure for pumping is developed, there is no longer much worry about breaking the carbon

37 27 foils. Just like going down to low pressures, however, going back up to atmospheric pressure presents the same issues and the same system of using small diameter pipes must be used. It is more desirable to let the system up to dry nitrogen so that water vapor does not get trapped or condense in difficult places, such as in the MCPs. Because of this, letting the system up to nitrogen enables a faster pump out back down to low pressures. Once the experimental area of the beam line is at the correct pressure, the experiment is ready to commence provided everything else is ready. The 100 Mo target used is one housed at ORNL which is 1.0 mg/cm 2 and of 97% purity. At this thickness, it can be considered a thick target, but thick targets are required for this type of low intensity system. If a thinner target is used, it would take far too long to obtain cross sections at a low level. The use of a thick target can also present problems due to energy loss in the target if one is working at sub-barrier energies, but this will be considered later. At a few points during the experiment, this 100 Mo target was exchanged for a 209 Bi target to get background spectra to be used later for data analysis. This bismuth target was 2 mg/cm 2 in thickness, and also of similar purity to the 100 Mo target. The targets in this experiment were then mounted on a frame that could be rotated to accommodate two different targets. One of these targets is, of course, the 100 Mo or 209 Bi target, but the other is a phosphor. The phosphor emits light when struck by energetic particles, which makes it much easier for the beam operators to tune the beam so that it is being delivered at the correct position. A small viewing window with a video camera, which is wired to the control room, peering in allows the operators to see exactly where they are delivering the beam. After the beam has been tuned, the 100 Mo or 209 Bi is returned to its position, ready to be bombarded by the beam.

38 28 Data Analysis Because there is no separation of the beam from the residues, data analysis can become a cumbersome task. To mitigate random background events, limits must be put on various parameters such as the beam register bit, time of flight, etc. To properly understand which parameters require limits and where to make them, the data acquisition system should be considered first. To get an in depth representation of the nuclear instrumentation modules used in this experiment, see figure 7, which shows the signal processing diagram. To summarize from before, a valid event is defined as either a scaled down beam event or a particle that has a slower flight time than the beam. Each time an event is registered, the Figure 7 Signal processing diagram of the data acquisition system. 26

39 29 type of event is recorded in the bit register, which operates in binary code. This bit register has a numeric value of 1, 2, or 3. A one indicates that the event was a scaled down beam event ( 01 in binary), a two indicates the event was slower than the beam ( 10 in binary), and a three (11 in binary) indicates that both events were present. In addition, the flight times between either the first and second or second and third timing detectors is recorded. These flight times are represented by the output of TAC (time-to-amplitude converter) modules. Other important parameters used in data analysis are signals from the three anodes of the ion chamber, which will be labeled de1, de2, and de3. These parameters just listed are the ones that are critical in off-line analysis. The other parameters recorded at the time of an event, such as the charges of the resistive layer and the individual MCP signals, are rarely used in off-line data analysis. These parameters are, however, very important during on-line analysis as they ensure that the system is functioning as it should be and that the beam is on target and at the proper intensity. As the experiment progresses, all of these parameters are initially stored in files with an.ldf extension. From here, it is a matter of preference as to what programs to use to read the data. However, different programs require different file types, so being able to read the data boils down to the conversion between file types. At ORNL, the data analysis programs used uses the files in their original.ldf format, but the analysis performed here is done using ROOT. 27 Each run of the experiment is comprised of many files, with a new file beginning once the prior one has reached 2 GB in size. In the conversion of the files to ROOT files, the file sizes drop to roughly half their size. What was originally 56 GB of data, in the.ldf format, is now about 20 GB in the form of.root files.